What is a Training Pattern in Machine Learning?
Training patterns are the input and target output values used to begin training a neural network, whether in a supervised or unsupervised role. While each node, or neuron, in a network has only one training pattern, there is no limit to how many nodes can be running different training patterns.
How are Training Patterns Used?
With supervised training, the desired inputs and outputs are provided by the trainer. The network then classifies the inputs and compares the resultant outputs against the benchmark outputs. Any errors are back-propagated throughout the system, which forces the network to adjust the various parameter weights. This continuous tweaking process repeats over and over, giving the “deep learning” name to the network.
With unsupervised training, the desired inputs are provided, but not the outputs. The neural network relies on various clustering and pattern recognition algorithms to group features for an output. This is also referred to as “adaptive learning.”